This proposal links the molecular biology of disease genes to the population-level processes that determine the frequencies of these genes and their evolutionary history. The host organism is the plant Arabidopsis thaliana and the pathogen is Pseudomonas viridiflava. The proposal focuses on host loci that affect resistance or susceptibility and pathogen loci that affect virulence (successful infection) or avirulence (unsuccessful infection). The first aim begins with measuring the phenotypic patterns of infection success or resistance for different host plants and pathogen isolates. In each of 10 populations, the investigators will analyze the outcome of infection for 50 pathogen isolates tested against each of 10 host genotypes. Next, three pathogen loci that affect virulence will be cloned. The three loci will be chosen based on their different infection successes when tested against host genotypes that have been well characterized at the sequence level and that provide helpful molecular tools for later analysis. Once the three pathogen loci have been chosen, the investigators will search for three matching loci in the host plant that interact with the pathogen virulence loci. Matching gene-for-gene interactions between plant and pathogen have been frequently observed and are likely to be found in this case. The final step for the first aim measures epidemiological aspects of natural populations. In particular, the investigators will study various natural populations for infection rates, host population densities and migration rates, and patterns of DNA polymorphism in the three host and pathogen loci that have been cloned. The second aim develops mathematical models. These models will be used to formulate hypotheses about how the population biology influences the frequencies of host and pathogen alleles and the pattern of molecular evolution at the cloned loci. The models emphasize how epidemiological processes of infection frequency and spread of disease influence aspects of gene frequency and molecular evolution. The third aim will use the data generated to estimate parameters of the mathematical models. Estimates include rates of epidemiological spread of disease, the fitness differences between plants that have or lack particular resistance alleles (cost of resistance), the fitness differences between pathogens that have or lack alleles that allow them to attack particular plant genotypes (costs of virulence), migration rates between populations, and the effects of environment (e.g., humidity) and host density on rates of pathogen transmission. The fourth aim uses the parameter estimates to expand the mathematical models and to study various aspects of epidemiology and evolution. For example, if weather has a significant impact on transmission, then that factor will be incorporated into the epidemiological components of the model. The model will be tested in the sense that the investigators will search for consistent explanations for how observed patterns of polymorphism, molecular evolution, and epidemiology fit together. For example, the epidemiology along with costs of resistance and virulence allow estimates for tendency of allele frequencies to fluctuate over time. The tendency of allele frequencies to fluctuate has, in turn, consequences for the expected patterns of molecular evolution. Thus, the investigators can use a particular aspect of their data to estimate processes such as the tendency of allele frequencies to fluctuate. They can then use their estimate of allele frequency fluctuations to make testable predictions about observable patterns, such as the distribution of nucleotide polymorphisms.